APPLIED AND ENVIRONMENTAL MICROBIOLOGY, June 1978, p. 1052-1060 0099-2240/78/0035-1052$02.00/0 Copyright © 1978 American Society for Microbiology

Vol..35, No. 6 Printed in U.S.A.

Phytoplankton Uptake and Excretion of Assimilated Nitrate in a Small Canadian Shield Lake Y. K. CHANt* AND N. E. R. CAMPBELL Department of Microbiology, University of Manitoba, Winnipeg, Manitoba, Canada Received for publication 2 December 1977

Nitrate uptake in the epilimnetic waters of a small eutrophic Canadian Shield lake was studied by using a "5N method during summer stratification. Concurrent with inhibition of primary production, 3-(3,4-dichlorophenyl)-1,1-dimethylurea inhibited NO3- assimilation. Nitrate up to 1 mg of N/liter did not affect the rate of primary production during 3 h of incubation. The N03 fertilizer added to the lake weekly was consumed through algal assimilation in about 3 days. Excretion of the photoassimilated N03 as dissolved organic nitrogen represented a significant portion of the nutrient incorporated by the cells. Only 40% of the N03-"5N which disappeared could be accounted for in the particulate fraction. Although the rest was presumably excreted, only 15% of the "5N label was accounted for as cationic dissolved organic nitrogen by isotope assays. These excreted organic forms were predominantly serine and glycine in the dissolved free amino acid fraction. Bacteria as well as algae might be expected to contribute to and modify the extracellular nitrogen pool.

Lake 227 is one of the 46 lakes of the Experimental Lakes Area in northwestern Ontario (15) first selected for long-term eutrophication studies. The lake measures 5 hectares in surface area, 10 m at maximum depth, and has had its phosphorus and nitrogen budgets controlled. Before artificial nutrient loading, Lake 227 was oligotrophic by all standards. From 1969 to 1974 inclusive, the lake was enriched with nitrogen as NO3- and phosphorus as P043- (6.29 g of N/m2 per year and 0.48 g of p/m2 per year) during the ice-free periods (24, 26). The immediate effect can be described as experimental eutrophication, and, to date, Lake 227 remains eutrophic. The phytoplankton standing crop was near the theoretical maximum (26). Ambient NO3- and P043- concentrations in Lake 227 remained low during summer stratification despite the artificial supply of these compounds, presumably due to phytoplankton consumption (26). Before the summer of 1975, when dissolved NO3- was distributed over the lake surface once a week, epilimnetic N03 concentration was reduced to barely detectable levels (ca. 10 ,ug of N per liter), and NO3- was not found below 5 m 2 days after its addition (unpublished data). It was therefore essential to assess the contribution of NO3- uptake to the release and cycling of nitrogen in this eutrophic system. Nitrate uptake in the epilimnetic waters,

which received nearly all of the fertilizers, had been observed to be much higher than that in the hypolimnion (unpublished data). The direct measurement of inorganic nitrogen uptake by "5N methodology was used in our investigation without precognition of extracellular excretion of assimilated NO3-. The kinetic approach had been introduced by Dugdale and Goering (5) who, assuming that excretion of the assimilated nitrogen was insignificant, performed experiments of a few hours of duration. The dependence of the rate of NO3- uptake on substrate concentration was studied in 1975 to characterize the algal population in Lake 227 with respect to NO3- assimilation. From these kinetic data the turnover time of NO3- (the time required for microbial uptake of the amount of NO3- equivalent to that supplied) was estimated. In addition, excretion of the photoassimilated NO3- as dissolved organic nitrogen (DON) was established and its significance was assessed. The extracellular DON was partially identified as dissolved free amino acids (DFAA) and combined amino acids. MATERIALS AND METHODS In the studies reported here, care was taken to clean all the apparatus and glassware by washing with dilute HCl and thorough rinsing with distilled-deionized water before use.

Sampling. Composite water samples were collected t Present address: Department of Microbiology, Macdon- between 0800 and 1000 h from a depth of 1.25 to 1.75 ald Campus of McGill University, Ste Anne de Bellevue, m at the center station on Lake 227 by using an opaque 3-liter Van Dorn bottle. These were stored in the dark Quebec, Canada HOA ICO. 1052

VOL. 35, 1978

N03- METABOLISM BY LAKE PHYTOPLANKTON

at 20°C for 2 to 4 before experiments were started. Light fixation of CO2 was discontinued during dark storage; at the start of the experiments the difference between light and dark uptake of NO3-, if any, could be accentuated. Nitrate uptake measurements. The "lightstarved" (dark-incubated) bulk samples were dispensed into 125-ml Pyrex reagent bottles, some of which were modified to provide lightproof conditions. Each replicate was then treated with 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) or P043- according to the particular experimental design after receiving a portion of K'5NO3 (95.0 atom %, International Nuclear & Chemical Corp.). Controls contained, in addition, 20 mg of HgCl2 per liter. Incubation was carried out in an illuminated water-bath incubator (7, 25) at 20°C ± 1°C and 6,285 J of illumination per m2 per min. After incubation, particulate nitrogen was retained on grade 984H Reeve Angel glass-fiber filters (which had been preignited at 500°C for 20 h) by vacuum filtration (125 mm Hg) and then determined as N2 by.dry combustion (20). The "5N abundance of each particulate nitrogen sample was determined by emission spectrometry on a Statron NOI-5 N-15 analyzer by the procedure developed by Flett (R. J. Flett, Ph.D. thesis, University of Manitoba, Winnipeg, 1976). Enrichment in '5N of the particulate fraction over that of background was taken to measure NO3- incorporation. Colorimetric determinations of N03- and N02- were also performed on the incubated water samples according to the method of Sawicki and Scaringelli (21), and that of NH3 was performed by the method of Stainton et al. (28). To study the effect of DCMU on the kinetics of N03 uptake, primary production was measured simultaneously by the 14C technique (29), but acidification and bubbling (27), instead of filtration, were employed to determine carbon uptake. Examination for extracellular products of nitrate assimilation. The problem of possible phytoplankton excretion of N03- assimilation products was investigated by the following experimental design to obtain unequivocal evidence for the existence of such excretion, to identify the chemical forms of these excretion products, and to estimate the net excretion rate under experimental conditions. The procedures used are summarized in Fig. 1 and described in detail below. At the time of the experiment, the phytoplankton in Lake 227 was predominantly made up of green algae (Spondylosium and Scenedesmus) and small Chrysophycean species. The test experiment consisted of a 4-liter epilimnetic water sample in an open cylindrical Pyrex jar. The sample was constantly mixed by means of a Teflon-coated stirring magnet driven by a motor-stirrer at a moderate rate and kept illuminated by two Sylvania Gro-Lux lamps installed 15 cm directly above. The fluorescent light source was enriched in the 425- and 650-nm regions, which together made up 35% of the total photo output. Its intensity was not measured but was lower (6,285 J/m2 per min) than that used in the N03 uptake experiments. At zero time, the sample was enriched with 100 yg of K`5N03N (95.0 atom %) per liter. During the 4-h incubation period at 22°C, the water temperature never rose more than 3°C, i.e., between 22 and 25°C. Control experiments also set up in parallel consisted of K`4N03-

1053

enriched or DCMU- or HgCl2-treated water samples similarly illuminated and a K'5NO3-enriched sample incubated in darkness. At the termination of incubation, subsamples were withdrawn for 5NO3- uptake measurements in the manner described above before the entire contents of each jar were continually fed to a Heraeus-Christ Junior 15000 centrifuge (ca. 100 ml/ min) to remove the large particulate matter at about 12,000 x g. All samples were centrifuged within 0.5 h of the end of the incubation time, and those awaiting centrifugation were kept under diffused light conditions to minimize continued photosynthetic activity. After centrifugation, 2.5 liters of each sample of the supernatant was consecutively filtered on preignited Whatman GF/C filters and membrane filters (Millipore Corp. HA, 0.45,um). The filtration pressure was maintained at 125 mm of Hg. Samples to be analyzed for amino acids and peptides, the suspected organic forms of excreted nitrogen, were first prepared from the filtrates by concentration on a Brinkman Rotovapor at 55°C to 1/100 of their original volumes. Fractions of the concentrates were used for peptide detection (13), protein estimation (17), and cation-exchange chromatography to isolate DON. Two resin columns, modified from 10-cm3 glass syringes and packed with Bio-Rad 50W-X12 (100 to 200 mesh, H form), were employed for each sample. Before the sample (concentrated filtrate) was applied to the first column, its pH was adjusted to about 2.0 with 0.1 HCI. After adsorption, the sample was washed with 40 ml of 0.1 N HCI, and the washing was applied to the second column. Each of the two columns was then washed with 40 ml of deionized water. Cations thus bound were then eluted with 1 M NH40H; a 25-ml portion of eluate was collected from each column and then pooled. Each pool sample was estimated for Lowry protein and divided into appropriate fractions according to protein content. These fractions were reduced to about 1-ml samples by rotary evaporation at 36°C on a Brinkmann Evapo-Mix before amino acid analyses and '5N determination were performed. The efficacy of the chromatographic procedure was tested by the separate application of known quantities of L-tyrosine, bovine serum albumin, calf-thymus deoxyribonucleic acid, and yeast ribonucleic acid (Sigma Chemical Co.) to the cation-exchange columns as described above. In each case, the test amino acid, protein, and nucleic acids were bound to the resin. Furthermore, recovery of these standards was close to 100% when the first 25 ml of eluate from each column was collected and assayed by absorption measurement at 260 or 280 nm. To remove the ammonia contained in the column eluates, these were first evaporated to dryness at 1100C and stored until amino acid analysis. The amino acid composition of acid-hydrolyzed and unhydrolyzed duplicate samples was determined by a Beckman 121 analyzer according to the method of Dexter and Dronzek (4). Fractions from the column eluates allocated for N isotope ratio determination were loaded onto preignited glass-fiber filters in a manner similar to that for the examination of glass-fiber filtrates. Traces of ammonia on the desiccated filters left from the column

1054

APPL. ENVIRON. MICROBIOL.

CHAN AND CAMPBELL

|INCUBATED |WATER SAMPLE_ Filter with ultrafifie GF filter

ColorimetricI

ContinuIOUs centrifulipation at 12,001e0 x g

PIATANT |

determinationof N03-, NO2 , and NH3

Filter consecutively with GF/C and 0.45pm membrane filters

|PART"ICULATE |FRW ,kCTION

,1;7--

Dry combusttion: particulate NI determination and Emission spe,ctrometry: N-isotope raltio analysis

2.5-liter

FILTRATE Concentrate 10(Ox by roto-evaporatio n

|CONCENTRATE| Peptide detection (ninhydrin) Protein estimation (Lowry)

Cation-exchange chromatography: isolation of organic N Elute with 1 M NH40H JELUATEI

Protein estimation (Lowry)

Evaporate to -1 ml

Evaporate to dryness

Dry combustion: organic N determiiination Acid hydrolysis

No

hydrolysis

and Emission spectrometry: N-isotope ratio analysis

Automatic amino acid analysis

FIG. 1. Summary ofprocedures used in the examination for extracellular products of NO3- assimilation.

elution procedure were removed when the filters were heated and evacuated in the sample preparation unit before "5N analysis. The amount of amino acids and DON and their '5N content were computed by knowing the original water filtrate volume represented by the sample analyzed.

RESULTS

Nitrate uptake kinetics. At an enrichment of 50 jig of N03 -N per liter, uptake was linear with time during 8 h of incubation on 23 August

1975. No lag period was observed. Nitrate uptake rates in the light with 1 and 10 ,uM DCMU and in the dark were in the ratio of 8:2:1:1. The fractional uptake rate of N03-, VNO,-, was 0.005 ,Lg of N/(microgram of total N) per h, or simply

0.005 h-', under light saturation. Dark uptake of N03 was small compared with that in the light. By using enrichment levels up to 500 ,ug of N03--N per liter on 5 September 1975, the light data displayed the Michaelis-Menten hyperbolic relationship between the rate of uptake and

N03- METABOLISM BY LAKE PHYTOPLANKTON

VOL. 35, 1978

substrate concentration (Fig. 2) and yielded the half-saturation transport constant, Kt, and the maximum fractional uptake rate, VmX,. by the Lineweaver-Burk, or double-reciprocal, plot (Fig. 3). These kinetic constants were determined to be about 73.6,ug of N per liter (or 5.2 ,uM) and 0.008 h-1, respectively. Although not shown here, Hofstee, or Eadie, plot (VNO:,- versus VNo,-/[NO3-]) and Hanes plot ([NO3-/ VNO:,- versus [NO3-]) of the same rectangular hyperbola also gave similar values of the kinetic constants. No Michaelis-Menten kinetic function was readily discernible from data obtained with the dark-incubated samples because of the low uptake rates. Therefore, these results were not amenable to linear analysis. Addition of 10 and 50 ug of P/liter seemed to enhance and inhibit N03 uptake, respectively (Fig. 2). However, unpublished data from studies of several lakes from the Experimental Lakes Area have shown phosphorus uptake with as high as 150jig of P per liter enrichment level for 24 h without any apparent toxic effect due to overloading (F. P. Healey, personal communication). Phytoplankton primary production was not affected by N03 enrichment up to 1 mg of N per liter (Table 1). Inhibition of primary production by DCMU was incomplete; at 1 ,uM, DCMU inhibited photosynthesis by 96% but still permitted CO2 fixation at a rate which was consistently two times higher than that of dark-incubated samples. Since an identical inhibition effect was noted in NO3- uptake experiments where 1 ILM DCMU was included (Fig. 4), it suggested that NO3- uptake is directly depend-

47 3-

NO;

,ug

N Litre

FIG. 2. Effect of N03- concentration on VNO, in Lake 227 epilimnetic waters (5 September 1975). Incubation was carried out for 4 h under the following conditions: I, light saturation without P043- addition; II and III, with 10 and 50 pg of P043--P per liter, respectively; Ia, IIa, and IIIa were similarly treated but incubated in the dark. See text for experimental details.

(NO3j',

ug

1055

N Litre

FIG. 3. Lineweaver-Burk linear transformation of N03- uptake kinetics data from Fig. 2. See legend to Fig. 2 for explanation of symbols.

ent on or related to electron transport to produce photoreductant. Extracellular products of NO3- assimilation. When results of l5NO3 incorporation were compared to those obtained by following the disappearance of NO3- chemically in the same experiments, a strong correlation (r = 0.99) was found between the two sets of data. However, uptake values obtained by "5N determination were consistently only about one-third of those obtained by colorimetric analyses (Fig. 5). Although the correlation coefficient r must vary when samples with widely different population composition and detritus nitrogen are studied, some preliminary results suggested that the discrepancy can also be due to the excretion of soluble organic nitrogen. First, neither N02 (nitrite) nor NH3 accumulation was detected during N03 uptake. Cell lysis was not likely to be significant during the short-term (3 or 4 h) incubation of freshwater samples. Second, preliminary electron microscopic examination of ultracentrifuge-precipitated materials from filtered water samples incubated with '5NO3 did not reveal any structural cell fragment. Hence, cell damage during filtration was not disclosed; little particulate nitrogen was recovered by ultracentrifugation of filtrates; and 15N was not incorporated into the sedimented materials. Third, control samples containing HgCl2 did not take up NO3- as determined by colorimetric or '5N analysis. Therefore, the colorimetric method of NO3- analysis was not at fault. It was already known from the N03 uptake results of light and dark experiments (Fig. 2 and 4) that this process was light dependent and was largely due to phytoplankton assimilation. From the preceding evidence, low-molecular-

1056 CHAN AND CAMPBELL

APPL. ENVIRON. MICROBIOL.

'1ABLE 1. Maximum primary production (Pnfmn with NO?,- enrichmenta Pnn,, (,uM C/h) with added NOV- at (/Ig of N/liter):

Mean

Pnmax,

Treatment

0 50 150 300 500 1,000 (,uM/ C/h) I. Light-saturated 2.95 NDb ND 2.98 2.89 2.73 2.89 II. With 1,uM DCMU ND 0.11 0.12 0.11 0.11 0.11 0.11 III. Dark-incubated 0.06 0.07 0.07 0.07 0.06 0.06 0.06 a N03- enrichment up to 1,000 jig of N per liter in Lake 227 epilimnetic waters (7 September 1975). Water samples were incubated for 3 h. See also Fig. 4. b ND, Not determined.

-- 35 1,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

z

z 30

6

Q

0

5-

.-

E 25-

4

o

I

xi.W

3

E II

2

0I

0

O

100

20304050

200

300 400

NO3

500 ug

0

0

0

600 700 800 N Litre

0

0

900 1000

FIG. 4. Effect of DCMU on the kinetics of N03uptake in Lake 227 epilimnetic waters (7 September 1975). Incubation was carried out for 3 h under the following conditions: I, light saturation; II, with 1 ,uM DCMU; III, in the dark. Maximum primary production was also estimated in the experiment by a 14C method. See also Table 1.

weight extracellular nitrogen was likely to be expected as a reasonable product to account for the observed N03 uptake discrepancies. The increase of '5N label in the DON isolated from filtrates of incubated water samples by cationexchange chromatography and the concurrent uptake of '5No3 are presented in Table 2. Increase in total particulate nitrogen after incubation was estimated to be less than 10%. Isotope analysis of total particulate nitrogen and DON demonstrated the cycling of N03 -N into DON. If the net DON production originated from the particulate fraction alone, about 20 to 45% of this DON in the light-incubated samples was primarily due to phytoplankton excretion. In the DCMU-treated and dark-incubated samples, 13 and 34%, respectively, of the DON released was due to phytoplankton and bacteria. It was noted that DCMU inhibited only 50% of N03 uptake in the DCMU experiment (20 July 1976, Table 2). A variation in the efficiency of inhibition by DCMU was therefore possible when different samples were used. DCMU appeared to inhibit DON production as well as N03 uptake. The

20/ 15

E/

10o C

10 0)

g

0 2 4 6 8 10 12 NO3 Uptake Determined By 15 N Method,,ug N Litre-I FIG. 5. Correlation between 15N tracer and colorimetric determnination of N03- uptake in Lake 227 epilimnetic waters.

net rate of fractional DON production was at least one-tenth as fast as that of fractional N03uptake; the rate of heterotrophic utilization of DON was unknown. As previously stated, nitrogen balances measured by colorimetric analysis and by particulate "5N determination showed a discrepancy of about 60% in terms of the total nitrogen applied. Of this, only about one quarter (15% of total) could be accounted for as DON by the isotope assay. Table 3 presents results from amino acid analyses of the isolated DON performed with and without acid hydrolysis. Analytical results obtained with acid treatment represented total amino acid content including both combined amino acids (peptides) and DFAA, whereas those from untreated samples represented DFAA only. Scrutiny of Table 3 discloses a

N03- METABOLISM BY LAKE PHYTOPLANKTON 1057

VOL. 35, 1978

TABLE 2. Net DONproduction from N03--'5N uptake in epilimnetic water samples (22 June and 20 July 1976J22 June

20 July

Determination Light

Dark

Light

1

IAM DCMU

Atom % '5N excess in DON 0.211 0.084 0.705 0.133 Net DON production (ug/liter) 0.16 0.06 0.66 0.12 Rate of DON production/h (x 103) 0.56 0.21 1.85 0.34 Atom % 15N excess in particulates 1.069 0.248 1.560 1.085 Net NO3- uptake (tg of N/liter) 6.23 1.44 9.46 5.37 Rate of N03- uptake/h (x 103) 2.81 0.65 4.10 2.85 a Incubation time, 4 h; total DON and total particulate nitrogen, ca. 80 and 560 ug/liter, respectively. TABLE 3. Amino acid composition of dissolved organic nitrogen isolated from filtrates of light-incubated water samples (20 July 1976)' A

Amino

acid'

Unhy-

Acid

hy-

Dif-

ferncec

drodrolyzed lyze lyzed lyzed

Try

Lys His Arg Asp Thr Ser Glu Pro Gly Ala Val Ile Leu Tyr Phe

3 54 2 0 33 18 92 27 15 60 34 11 11 16 6 3

NDd 67 3 4 63 37 98 110 30 138 100 33 23 33 7 24

C

B

ND 13 1 4 30 19 6 83 15 78 66 22 12 17 1 21

Unhy-

hydrolyzd dro dro

~lyzed 3 ND 46 2 1 29 19 81 34 19 52 39 8 9 15 3 3

74 5 5 58 30 91 135 30 151 124 24 22 36 6 22

D

Dif Unh - Aci Unhy- Aci cDiifd drohydroferdro id ence droe ence'.do lyzed drolyzed droe lyzed lyzed

fer-

ND 28 3 4 29 11 10 101 11 99 85 16 13 21 3 19

2 31 0 0 20 11 51 21 15 41 24 6 7 11 0 0

ND 52 2 2 51 32 54 73 20 114 70 24 17 24 4 6

ND 21 2 2 31 21 3 52 5 73 46 18 10 13 4 6

6 20 0 0 11 12 52 15 12 40 25 5 4 11 0 0

ND 36 2 0 35 26 54 54 25 105 52 15 11 18 2 5

Dif

ference

ND 16 2 0 24 14 2 39 13 65 27 10 7 7 2 5

385 770 388 363 Total 813 453 240 545 307 213 440 233 a Treatments: A, 100 yg of '5NO3--N per liter enrichment; B, 100 Lg of '4NO3--N per liter enrichment; C, 100 Lg of `5N03-N per liter and 1 liM DCMU added; D, 100 yg of '5NO3-N and 20 mg of HgCl2 per liter added. Results are expressed in nanomoles per liter of water sample. b Cysteine and cystine were not determined. Tryptophan was destroyed by acid hydrolysis. Traces of methionine are not included. c Difference between acid-hydrolyzed and unhydrolyzed samples. d ND, Not determined.

convenient partition of the data into two groups, namely, A and B versus C and D (controls). Each group has common DFAA and peptide concentrations that are significantly different from those of the other. A comparison of the two groups showed two prominent features. First, light-incubated samples enriched with either '5NO3 or '4NO3 were about 150 nM higher in total DFAA concentration than control samples treated with DCMU or HgCl2. Treatment with HgCl2 did not promote cell lysis because there was no significant increase in total DFAA or combined amino acids over background levels. Second, most individual DFAA concentra-

tions were low (usually